CN114585889A

CN114585889A - Method and system for element identification by optical emission spectroscopy

Info

Publication number: CN114585889A
Application number: CN201980101492.4A
Authority: CN
Inventors: D·F·麦卡锡; M·A·伍德
Original assignee: Agilent Technologies Inc
Current assignee: Agilent Technologies Inc
Priority date: 2019-10-22
Filing date: 2019-12-23
Publication date: 2022-06-03
Also published as: WO2021079184A1; JP2022553352A; US20240053201A1; KR20220080194A; EP4048991A1; EP4048991A4

Abstract

The present invention relates to a computer-implemented method of automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy. The method comprises the following steps: obtaining sample spectral data from the sample; obtaining, for each element of the periodic table of elements that is quantifiable by optical emission spectroscopy, a list of one or more predetermined emission wavelengths, each predetermined emission wavelength associated with a list of one or more potentially interfering emission wavelengths; determining a list of one or more analyte wavelengths corresponding to spectral peaks in the sample spectral data based on the list of emission wavelengths; determining, for each analyte wavelength, whether the corresponding spectral peak has a likelihood of being affected by an interfering emission wavelength that causes spectral interference based on the list of one or more potential interfering emission wavelengths corresponding to the analyte wavelength; determining a revised list of one or more analyte wavelengths by removing from the list of analyte wavelengths that correspond to spectral peaks having a likelihood of being affected by interfering emission wavelengths; and determining a confidence level that one or more elements are present in the sample based on a set of criteria applied to the revised list of analyte wavelengths.

Description

Method and system for element identification by optical emission spectroscopy

Technical Field

The present invention relates to a method and system for element identification by optical emission spectroscopy, such as inductively coupled plasma optical emission spectroscopy (ICP-OES) (also known as inductively coupled plasma atomic emission spectroscopy (ICP-AES)), although the scope of the invention is not necessarily limited thereto.

Background

Spectrometric techniques are used to identify the presence of a target chemical or analyte in a sample. Some spectrometric techniques rely on the interaction of an analyte with an excitation source (e.g., light) in the visible spectrum or in invisible wavelengths. Depending on the particular spectrometric technique employed, the collected spectra may show the intensity of light absorbed or emitted by the sample after interaction of the light beam with the sample.

In other spectrometry techniques, the excitation source is a plasma source, typically made of argon, which provides plasma energy to the atomized sample causing the constituent atoms to be excited and emit light. The emitted light is directed into the spectrometer via an entrance slit or aperture that limits the amount of light that enters the system. The optical device disperses light entering the system to separate different wavelengths of the emission spectrum. The detector simultaneously records multiple wavelength ranges to capture the emission of multiple elements in different parts of the emission spectrum. The detector is typically an integrated array of photosensitive elements for collecting light passing through the spectrometry system. The spatial separation of the individual spectra on the array detector is used to distinguish between different wavelengths of light absorbed or emitted by the sample.

Peaks or troughs in the spectral profile of the detected light indicate the presence of a particular chemical in the sample. In some spectrometric techniques, the amount or relative amount of each chemical can thus be derived from the spectrum.

Conventional methods in ICP-OES typically require users to accurately specify the elements they wish to quantify in any given analytical method. This method requires the operator to know in advance which elements in the sample are of interest.

Typically, an operator will obtain a list of elements from a pre-existing prescribed method that will include all elements deemed likely to be of interest at the time of method verification.

This method is very limited because the operator cannot evaluate the composition of the sample except for the elements specified in the method. This may result in important sample components being ignored, particularly in anomalous samples, which may contain components that are rarely found in homogeneous samples.

In addition to requiring users to specify elements of interest to them, existing methods in ICP-OES require users to specify one or more wavelengths for quantification of any given element. Typically, the wavelength must be specified before the analysis begins. In some cases, wavelengths may be subsequently assigned when a sufficiently large spectral region containing many wavelengths for analysis is collected during the measurement. However, both of these cases require an experienced user to select a wavelength appropriate for the sample being analyzed.

The quantitative selection of appropriate wavelengths for elements in ICP-OES is complicated and incorrect wavelength selection is a common source of error for this analytical technique. There are some documented effects that when quantifying an element by ICP-OES, it may cause any given wavelength to give incorrect results.

Spectral and non-spectral interference are two means by which the intensity of an elemental measurement in a sample solution at any given analytical wavelength can be affected and return a potentially incorrect result.

Non-spectral interference occurs when constituents in the sample affect sample introduction or plasma conditions, and can result in the enhancement or suppression of one or more analytical wavelengths of a given element.

Spectral interference occurs when the analytical wavelengths are partially or completely overlapped by the emission of another element or molecule, or are affected by unstructured background radiation. The presence and magnitude of spectral interference is highly sample dependent, even in the same method.

One current method of avoiding spectral interference in a sample is to select an analysis wavelength that is suspected of not interfering with the sample being analyzed. This process is entirely manual, relying on the experience and knowledge of the operator operating the instrument. To ensure that all samples achieve effective results, the operator will typically include multiple wavelengths of the same element in the process. However, this method provides the user with multiple results for each element in each sample, which must then be interpreted to determine which is the most appropriate.

Some alternatives do not attempt to avoid spectral interference, but rely on various calibration techniques, such as inter-element correction (IEC), to correct for their effects.

These methods have been widely accepted and provide reasonably accurate results. However, they rely on the instrument operator to correctly identify the interfering elements and then prepare the appropriate chemical standards so that interference corrections can be calculated and applied. Thus, these current methods require the operator to know in advance the interfering elements that may be present in their sample. When interferents are not obvious or common, or the operator is analyzing a sample type that they are not familiar with, it is often difficult for the operator to successfully predict the presence of interfering elements.

Embodiments of the present invention may provide a method and system for element identification by optical emission spectroscopy that overcomes or ameliorates one or more of the disadvantages or problems described above, or at least provides the consumer with a useful choice.

Disclosure of Invention

According to one aspect of the present invention, there is provided a computer-implemented method of automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy, the method comprising the steps of:

obtaining sample spectral data from the sample;

obtaining, for each element of the periodic table of elements that is quantifiable by optical emission spectroscopy, a list of one or more predetermined emission wavelengths, each predetermined emission wavelength associated with a list of one or more potentially interfering emission wavelengths; determining a list of one or more analyte wavelengths corresponding to spectral peaks in the sample spectral data based on the list of emission wavelengths;

determining, for each analyte wavelength, whether the corresponding spectral peak has a likelihood of being affected by an interfering emission wavelength that causes spectral interference based on the list of one or more potential interfering emission wavelengths corresponding to the analyte wavelength;

determining a revised list of one or more analyte wavelengths by removing from the list of analyte wavelengths that correspond to spectral peaks having a likelihood of being affected by interfering emission wavelengths; and

determining a confidence level that one or more elements are present in the sample based on a set of criteria applied to the revised list of analyte wavelengths.

Advantageously, the revised list of analyte wavelengths is substantially pre-processed prior to the step of determining the presence of any elements in the sample to remove those analyte wavelengths that may be susceptible to spectral interference. In this manner, the computer-implemented method can be used by an operator for any uncharacterized sample without requiring any experience or knowledge about the sample or the associated instrument work.

In one embodiment, the sample spectral data comprises data representing emission intensities corresponding to wavelengths within a spectral range of the sample.

Sample spectral data from a sample can be obtained using any suitable analyzer in optical emission spectroscopy. For example, optical emission spectroscopy may be used to obtain sample spectral data. In particular, an ICP-OES or ICP-AES instrument may be used to obtain sample spectral data from a sample.

In one embodiment, the analyzer interfaces with a computer having a processor. The interface may be a wired or wireless connection. The computer processor may include a software application installed thereon for performing one or more steps of a computer-implemented method. In an alternative embodiment, the software application may be a cloud-based application accessible via a network such as the internet. In some embodiments, the software application may be accessed remotely via a local network.

Typically, a list of one or more predetermined emission wavelengths is compiled based on standard emission wavelength measurements of each element made for a particular type of analyzer (e.g., an ICP-OES instrument).

A list of one or more potentially interfering emission wavelengths may be determined based on the proximity of neighboring emission wavelengths to each standard emission wavelength (associated with a particular element). Adjacent emission wavelengths may be associated with different elements of a particular element and cause spectral interference in the intensity measurements of the particular element at the emission wavelengths.

A list of one or more predetermined emission wavelengths and associated potentially interfering emission wavelengths may be stored in a database in computer memory. In some embodiments, the database may be stored remotely and may be accessed over a network or the internet. The database may be accessed during execution of a software program to implement the steps of the computer-implemented method.

Further, the step of determining the list of analyte wavelengths may comprise: analyzing a region of interest of the spectral range of the sample corresponding to each predetermined emission wavelength of each element, and determining whether a peak of emission intensity is located within the region of interest.

In general, if a peak in emission intensity is located, the emission wavelength corresponding to the peak of the localized spectrum of the element in the sample is referred to herein as the analyte wavelength corresponding to the element in the sample.

In particular, the step of determining the list of analyte wavelengths may comprise: analyzing a region of interest of the sample spectral range corresponding to each predetermined emission wavelength for each element, determining whether a saturation result is located within the region of interest, and upon determining that the saturation result is not located within the region of interest, determining whether a peak in the emission intensity is located within the region of interest.

The saturation results may be associated with emission intensity measurements that are outside the measurement range of the analyzer instrument. In general, saturation results may be encountered when there is spectral interference and/or high concentrations of the corresponding elements in the sample.

The step of determining the list of analyte wavelengths may further comprise determining whether the saturation result represents a peak in emission intensity with a flat top.

The step of determining the list of analyte wavelengths may be performed by a software program installed on the processor using information from the database about the emission and interference wavelengths.

Additionally, the step of determining the list of analyte wavelengths may further comprise determining a confidence level that a peak in the emission intensity has been identified in the region of interest based on a threshold test.

Any suitable means may be used to indicate the confidence level. For example, the confidence level may be represented by a confidence factor. The confidence factor may be expressed as a numerical value within a predetermined range.

In some embodiments, determining a confidence level that a peak in the emission intensities has been identified in the region of interest may include calculating a standard deviation of the emission intensities in a vicinity of the peak to determine a confidence factor.

The step of determining the confidence level may be performed by a software program installed on the processor.

In some embodiments, elements associated with a peak may be considered identified if the confidence factor is greater than a predetermined threshold. The predetermined threshold may be calculated based on historical and/or experimental sample data. The predetermined threshold may also be adjusted according to the particular type of analyzer instrument or the particular instrument based on experimental data collected using the instrument.

In some embodiments, the step of determining whether the corresponding spectral peak for each analyte wavelength has a likelihood of being affected by an interfering emission wavelength may comprise the steps of:

determining a clean interfering emission wavelength associated with each analyte wavelength, an

Determining whether the clean interfering emission wavelength corresponds to a spectral peak in the sample spectral data.

The step of determining whether the corresponding spectral peak for each analyte wavelength has a likelihood of being affected by an interfering emission wavelength may be performed by a software program installed on a processor.

The clean or cleanest interfering emission wavelength generally refers to the interfering emission wavelength that is least likely to be affected by the spectral interference itself, of all the potentially interfering emission wavelengths associated with the analyte wavelength. Furthermore, the cleanest interfering emission wavelength may be the dominant emission wavelength associated with the corresponding interfering element, thus providing a clear intensity measurement if the interfering element is detected in the sample.

In some embodiments, the step of determining a clean interfering emission wavelength may include determining an interfering emission wavelength that is least likely to be affected by spectral interference.

In some embodiments, the method may further comprise

Determining, for each analyte wavelength corresponding to a spectral peak affected by spectral interference, a significance of the spectral interference based on any one or more of:

a distance between a spectral peak corresponding to the clean interfering emission wavelength and a spectral peak corresponding to the associated analyte wavelength;

a ratio of a spectral peak corresponding to the clean interfering emission wavelength and a spectral peak corresponding to the associated analyte wavelength; and

a ratio of an emission intensity corresponding to the clean interfering emission wavelength to an emission intensity corresponding to the associated analyte wavelength.

The emission intensity may be predetermined in terms of the intensity of the spectral lines.

The step of determining the significance of the spectral disturbance emission wavelength may be performed by a software program installed on a processor.

In one embodiment, the set of criteria for determining the confidence level that one or more elements are present in the sample may include any one or more of:

whether a number of detected primary analyte wavelengths corresponding to each element in the revised list of analyte wavelengths is above a first threshold; and

whether a number of detected primary and secondary analyte wavelengths corresponding to each element in the revised list of analyte wavelengths is above a second threshold,

wherein the primary analyte wavelength of the element corresponds to an emission wavelength having a high peak spectral intensity and the secondary analyte wavelength of the element corresponds to an emission wavelength having a peak spectral intensity lower than the peak spectral intensity of the primary analyte wavelength.

The first and second thresholds may be determined based on a desired minimum confidence level. The desired minimum confidence level may be determined based on user requirements, industry standards, and/or regulatory requirements.

In one embodiment, the first threshold is two for elements having at least three primary analyte wavelengths, and the first threshold is one for elements having two or less primary analyte wavelengths, and the second threshold is at least one primary analyte wavelength and one secondary analyte wavelength.

The computer-implemented method may further include adding one or more elements to the list of identified elements based on the determined confidence level. The list of identified elements may be stored in a memory of the computer device.

In some embodiments, after performing the computer-implemented method, each analyte wavelength is ranked according to a confidence factor that indicates a confidence level that the corresponding analyte element is present in the sample. For example, analyte wavelengths that are less likely to be affected by spectral interference may be given a relatively high confidence factor, while analyte wavelengths that are more likely to be affected by spectral interference may be given a relatively low confidence factor.

In some embodiments, if the number of detected primary analyte wavelengths corresponding to each element in the revised list of analyte wavelengths is above a first threshold, those detected primary analyte wavelengths may be assigned a first confidence factor. The first confidence factor may be a relatively high confidence factor. The corresponding analyte element may also be assigned a first confidence factor.

Further, if the number of detected primary and secondary analyte wavelengths corresponding to each element in the revised list of analyte wavelengths is above a second threshold, those detected primary and secondary analyte wavelengths may be assigned a second confidence factor. The corresponding analyte element may also be assigned a second confidence factor. The second confidence factor is lower than the first confidence factor.

If the number of detected primary analyte wavelengths corresponding to each element in the revised analyte wavelength list is below a first threshold and the number of detected primary and secondary analyte wavelengths corresponding to each element in the revised analyte wavelength list is below a second threshold, those detected primary and secondary analyte wavelengths may be assigned a third confidence factor. The corresponding analyte element may also be assigned a third confidence factor. The third confidence factor is lower than the second confidence factor.

In some embodiments, the elements in the list of identified elements are ordered according to their associated confidence factors.

In some embodiments, the computer-implemented method may further comprise

Validating each element in the list of identified elements to determine whether a spectral peak of the sample spectral data associated with an analyte wavelength is likely to be affected by an interfering emission wavelength causing spectral interference, and

upon determining that an analyte wavelength having a corresponding element in the list of identified elements is likely to be affected by an interfering emission wavelength that causes spectral interference, removing the corresponding element from the list of identified elements.

The step of verifying each element may be performed by a software program installed on the processor.

Advantageously, the step of validating each element provides an opportunity for each element in the list of identified elements to be re-evaluated, such that any elements found in the sample and added to the list of identified elements that may be misqualified are removed.

The computer-implemented method may further include selectively displaying an analyte wavelength corresponding to each element in the list of identified elements based on selection criteria. The selection criteria may include any one or more of:

whether the analyte wavelength is associated with a saturation result,

maximum number of analyte wavelengths displayed for each corresponding element, and

whether the analyte wavelength is associated with a user selection.

The computer-implemented method may further include calculating a concentration for each element in the list of identified elements. The step of calculating the concentration of each element may comprise measuring the emission intensity of a spectral peak associated with the corresponding element and correcting for background emission.

The computer-implemented method may further include identifying an anomalous analyte wavelength and reducing the confidence level that a corresponding element is present in the sample based on a measurement associated with the anomalous analyte wavelength. In some embodiments, the confidence level that the corresponding element is present in the sample is inferred based on the confidence level that the analyte wavelength associated with the element is detected.

According to another aspect of the present invention there is provided a system for automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy, the system comprising:

an optical emission spectrometer for obtaining sample spectral data from the sample; and

a processor for performing the computer-implemented methods described herein.

According to another aspect of the invention, there is provided one or more tangible, non-transitory computer-readable media having computer-executable instructions for performing a computer-implemented method as described herein.

According to a further aspect of the present invention there is provided a computer system for automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy, the system comprising:

a sample data retrieval module for obtaining sample spectral data from the sample;

a wavelength data retrieval module for obtaining, for each element of the periodic table of elements quantifiable by optical emission spectroscopy, a list of one or more predetermined emission wavelengths, each predetermined emission wavelength associated with a list of one or more potentially interfering emission wavelengths;

a peak search module for determining a list of one or more analyte wavelengths corresponding to spectral peaks in the sample spectral data based on the list of emission wavelengths;

an interference search module to determine, for each analyte wavelength, whether the corresponding spectral peak has a likelihood of being affected by an interfering emission wavelength that causes spectral interference based on the list of one or more potential interfering emission wavelengths corresponding to the analyte wavelength;

a wavelength processing module that determines a revised list of one or more analyte wavelengths by removing from the list of analyte wavelengths that correspond to spectral peaks that have a likelihood of being affected by interfering emission wavelengths; and

an element identification module for determining a confidence level that one or more elements are present in the sample based on a set of criteria applied to the revised list of analyte wavelengths.

The sample data retrieval module may interface with an optical emission spectrometer to obtain sample spectral data from a sample.

The wavelength data retrieval module may retrieve the wavelength data from a database.

The peak search module may be configured to

Analyzing a region of interest of the spectral range of the sample corresponding to each predetermined emission wavelength of each element, determining whether the saturation result is located within the region of interest, and

upon determining that the saturation result is not located within the region of interest, determining whether a peak in the emission intensity is located within the region of interest.

The peak search module may be further configured to determine whether the saturated result represents a peak in the emission intensity with a flat top.

The peak search module may be further configured to determine a confidence level that a peak in the emission intensity has been identified in the region of interest based on a threshold test. Determining a confidence level that a peak in the emission intensities has been identified in the region of interest may include calculating a standard deviation of emission intensities in a vicinity of the peak to determine a confidence factor.

The interference search module may be further configured to determine a clean interfering emission wavelength associated with each analyte wavelength and determine whether the clean interfering emission wavelength corresponds to a spectral peak in the sample spectral data.

The computer system may be further configured to determine, for each analyte wavelength corresponding to a spectral peak affected by spectral interference, the significance of the spectral interference based on any one or more of:

a ratio of the emission intensity corresponding to the clean interfering emission wavelength and the emission intensity corresponding to the associated analyte wavelength.

The element identification module may be configured to determine a confidence level that one or more elements are present in the sample based on any one or more of:

The element identification module may be configured to add one or more elements to the list of identified elements based on the determined confidence level.

The computer system may further include a validation module for validating each element in the list of identified elements to determine whether a spectral peak of the sample spectral data associated with the respective element is likely to be affected by a disturbing emission wavelength causing spectral interference, and upon determining that an element in the list of identified elements is likely to be affected by a disturbing emission wavelength causing spectral interference, removing the corresponding element from the list of identified elements.

The computer system may further include a result selection module for selectively displaying analyte wavelengths corresponding to each element in the list of identified elements based on selection criteria, wherein the selection criteria include any one or more of:

whether the analyte wavelength is associated with a saturation result,

whether the analyte wavelength is associated with a user selection.

The computer system may further include a concentration calculation module to calculate a concentration for each element in the identified list of elements. Calculating the concentration of each element may include measuring an emission intensity of a spectral peak associated with the corresponding element and correcting for background emission.

The computer system may further include an anomaly detection module to identify an anomalous analyte wavelength and to reduce the confidence level that a corresponding element is present in the sample based on a measurement associated with the anomalous analyte wavelength.

Advantageously, embodiments of the present invention automatically identify elemental emission wavelengths in an uncharacterized solution to identify all elements that may be available for ICP-OES technology without requiring an operator to pre-select the elemental emission wavelengths.

In order that the invention may be more readily understood and put into practical effect, one or more preferred embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings.

Drawings

FIG. 1a is a schematic diagram illustrating a system for element identification using optical emission spectroscopy, according to one embodiment of the present invention. FIG. 1a further illustrates a process flow diagram outlining the process steps of a computer-implemented method for element identification in accordance with one embodiment of the present invention.

FIG. 1b illustrates display information related to background emission correction and concentration calculation in the computer-implemented method of FIG. 1 a.

FIG. 2a is a process flow diagram illustrating the data acquisition sub-process of the computer-implemented method shown in FIG. 1 a.

FIG. 2b is a process flow diagram illustrating a data loading sub-process of the computer-implemented method shown in FIG. 1 a.

FIG. 2c illustrates display information provided by the sub-process shown in FIG. 2b relating to standard transmit wavelength data and associated potentially interfering transmit wavelength data.

FIG. 3 is a process flow diagram illustrating an element search process of the computer-implemented method shown in FIG. 1 a.

FIG. 4 is a process flow diagram illustrating a sub-process of determining spectral peaks in the element search process shown in FIG. 3.

Fig. 5a shows display information according to an exemplary embodiment of the present invention relating to the concentration results determined for some of the identification elements of samples 1 to 10 according to the method shown in fig. 1 a.

Fig. 5b shows displayed information relating to further results, including a graphical representation of the analyte wavelength, the corresponding confidence rating and the spectral data of the analyte wavelength of the element lithium (Li) in sample 1 shown in fig. 5 a.

Fig. 5c shows display information relating to the concentration results determined for some of the identification elements of samples 1 to 10 according to the method shown in fig. 1a, according to an exemplary embodiment of the present invention.

Fig. 5d shows displayed information relating to further results, including a graphical representation of the analyte wavelength, the corresponding confidence rating, and the spectral data for the analyte wavelength of elemental lithium (Li) in sample 7 shown in fig. 5 c.

Fig. 6 is a process flow diagram illustrating a sub-process of determining an interference spectrum peak in the element search process shown in fig. 3.

Fig. 7a shows display information relating to the concentration results determined for some of the identification elements of samples 1 to 10 according to the method shown in fig. 1a, according to an exemplary embodiment of the invention.

Figure 7b shows displayed information relating to further results including analyte wavelength, corresponding confidence rating and a graphical representation of spectral data for analyte wavelength in sample 5 as shown in figure 7 a.

Fig. 7c shows display information relating to the concentration results determined for some of the identification elements of samples 1 to 10 according to the method shown in fig. 1a, according to an exemplary embodiment of the present invention.

Fig. 7d shows displayed information relating to further results, including analyte wavelengths, corresponding confidence ratings, and a graphical representation of spectral data for analyte wavelengths in the sample 10 as shown in fig. 7 c.

FIG. 8 is a process flow diagram illustrating a sub-process for determining the presence of an analyte element in the element search process shown in FIG. 3.

FIG. 9 is a flow chart illustrating a process of verifying and re-evaluating the presence of spectral interference in the method shown in FIG. 1 a.

FIG. 10 is a flow chart illustrating a process for selectively determining acceptable analyte wavelengths for display in the method shown in FIG. 1 a.

FIG. 11 is a process flow diagram illustrating a process for determining outlier results in the method shown in FIG. 1 a.

FIG. 12 is a flow chart illustrating a process of selecting the best available result for display in the method shown in FIG. 1 a.

FIG. 13 shows displayed information in the form of a graphical representation of identified analyte wavelengths and analyte elements in a selected portion of sample spectral data according to an example embodiment of the present invention.

FIG. 14 is an excerpt of a user interface of the system according to an embodiment of the invention, illustrating the automatic identification of several common and problematic spectral disturbances of the wavelengths of As, Mn, and V in the HJ781-2016 solid waste digest.

Fig. 15 is an excerpt of a user interface of a system according to one embodiment of the invention, illustrating that no Cl is detected in "soil 4" and no Sb is in "soil 4" due to the carelessness of the technician when adding acid prior to digestion. This may not be dissolved in the sample due to the lack of HCl in the digest.

Fig. 16a and 16b are excerpts of a user interface of a system comparing visualizations of periodic table heat patterns of a plurality of samples according to an embodiment of the invention. In this embodiment, concentration-based color coding may provide a visually intuitive way to identify differences between measurement solutions.

FIG. 17 is an excerpt of a user interface of a system according to one embodiment of the invention showing a confidence rating table for Mn analyte wavelengths. The user interface also provides an information box for displaying possible Fe interference on the two Mn main lines according to the user request.

Fig. 18 further illustrates a heat map showing the relative concentrations of all other elements in the sample.

Detailed Description

As shown in FIG. 1a, a system 100 for automatically identifying the presence of one or more elements in a sample by optical emission spectroscopy includes an optical emission spectrometer 102, such as an inductively coupled plasma optical emission spectroscopy (ICP-OES) instrument, also known as inductively coupled plasma atomic emission spectroscopy (ICP-AES). The ICP-OES instrument 102 obtains spectroscopic data from one or more samples for analysis. The system 100 further includes a processor (not shown) having an application installed thereon for executing a software-implemented method 106 of analyzing sample spectral data obtained from the instrument 102 and identifying the presence of one or more elements in the sample with a level of confidence. In some cases, if the elements in the sample are below the detection level, the instrument 102 may not be able to identify the presence of any elements. A display device 104 is coupled to the processor for providing a user interface to facilitate user interaction with the system 100 and displaying output from the sample analysis.

The computer-implemented method 106 obtains sample spectral data from one or more uncharacterized sample solutions loaded into the instrument 102 and automatically analyzes the samples in a series of functional steps to identify the presence of one or more elements in each sample, as described in further detail below. The method 106 is described with respect to a single sample. However, it should be understood that the method 106 is not limited to analyzing a single sample and can process any suitable number of samples.

At start step 200, sample spectral data is acquired from a sample solution loaded into the instrument 102 and theoretical emission wavelength data is loaded from a data repository. The data repository provides a list of all elements in the periodic table that can be quantified by optical emission spectroscopy (also referred to herein as an element list) and theoretical emission wavelength data for each element in the element list. Theoretical emission wavelength data may be compiled manually based on measurements of standard samples and/or based on standard data, such as those in atomic spectroscopy databases published by the National Institute of Standards and Technology (NIST). In particular, the theoretical emission wavelength data includes a standard emission wavelength for each element in the list of elements, and potential interference wavelength data associated with each standard emission wavelength. More details about the step initiation step 200 will be described in more detail below with reference to fig. 2a to 2 c.

At query step 108, the method 106 checks whether the element search process 300 has been performed for each element in the list of elements for the sample solution under analysis. If so, the method 106 continues to process 700. If not, the method 106 continues with performing the process 300 for the next element in the list of elements.

In summary, the process 300 compares the theoretical emission wavelength data to sample spectral data from the sample solution to determine a list of analyte wavelengths that may be free of spectral interference, which correspond to spectral peaks in the sample spectral data. The process 300 then further determines the confidence level that one or more elements are present in the sample by evaluating the list of analyte wavelengths against a predetermined criterion. The process 300 generates a list of elements that may be present in the sample based on predetermined criteria and a list of analyte wavelengths used to identify these elements (the identified elements and the list of analyte wavelengths 110). Process 300 is an iterative process that is applied to each element in the list of elements. The process 300 is described in more detail with reference to fig. 3.

At process 700, method 106 validates and reevaluates the list of identified elements and analyte wavelengths 110 to determine any further spectral interference. Any analyte wavelengths from the list 110 that are determined to be affected by spectral interference are removed from the list 110 and/or given a low confidence rating. Process 700 is described in further detail below with reference to fig. 9.

At process step 800, the method 106 further evaluates the list of identified elements and analyte wavelengths 110 and selects the most appropriate analyte wavelength for each element in the list 110 for display on the display device 104.

At step 112, for each analyte wavelength in list 110, a background emission correction is applied to determine a net spectral peak intensity associated with the analyte wavelength. Any standard ICP-OES background correction technique can be used. For example, due to its robustness, a fitting background correction technique may be used. In step 112, for each analyte wavelength in list 110, a semi-quantitative concentration of the relevant element is calculated using a predetermined intensity-concentration calibration curve. Any standard ICP-OES calibration curve may be used, for example the calibration curve may be linear or quadratic.

For example, as shown in FIG. 1b, the display information generated by method 106 includes a list of analyte wavelengths associated with elemental sodium (Na) identified in the sample. The 589.892nm first analyte wavelength 116 for Na had a 3 star confidence rating 118, a calculated concentration of 0.51mg/L, an intensity of 7874.7c/s, and a background emission of 36933.7 c/s.

Plot 126 of fig. 1b is a plot of intensity versus wavelength and shows a spectral peak 128 in the sample spectral data for Na analyte wavelength 589.892nm in comparison to an estimate of background emission 130.

Similarly,

graphs

132 and 134 show portions of sample spectral data associated with Na analyte wavelengths 588.995nm and 568.821nm, respectively.

Referring now to FIG. 1a, at process step 900, the method 106 evaluates the analyte wavelengths in the list 110 to obtain possible outlier results for each element and adjusts the confidence rating associated with the corresponding analyte wavelength. The process 900 is described in more detail with reference to fig. 11.

At process step 1000, method 106 weights the analyte wavelengths in list 110 based on weighting criteria to identify the most appropriate results for display. The process 900 is described in more detail with reference to fig. 12.

Referring now to fig. 2a and 2b, there are shown

subprocesses

202, 212 associated with the sample data acquisition and database loading step 200 of the method 106.

During sub-process 202, sample spectral data is obtained from an uncharacterized sample solution using instrument 102.

At a start step 204, an uncharacterized sample solution is loaded into the instrument 102. Although the present specification describes the process with respect to a single sample solution, those skilled in the art will appreciate that the instrument is generally configured to sequentially retrieve spectroscopic data from a plurality of sample solutions.

At step 206, the instrument 102 obtains spectroscopic sample data from the uncharacterized sample solution. The spectroscopic sample data provides measured intensities of wavelengths within the spectral range of the sample. Typically, the sample spectral data will consist of many data points covering a wide range of wavelengths and signal intensities.

The output 208 from step 206 is stored in a data repository 210 of the system 100 for use in the method 106 and is referred to herein as sample spectral data.

During subprocess 212, standard atomic emission wavelength data is loaded from data repository 218 for use during execution of method 106.

At query step 214, the sub-process 212 checks whether standard atomic emission wavelength data has been retrieved for all elements in the list of elements. If so, the sub-process 212 is complete, and if not, the sub-process 212 proceeds to step 216 for each remaining element. Sub-process 212 is an interactive process and the steps described below are applied to each element in the list of elements until all elements in the list of elements have been loaded with all relevant emission wavelength data.

At step 216, the sub-process 212 retrieves the first ten primary element emission wavelengths from the data repository 218 and stores the emission wavelengths in the list 222.

At step 220, for each emission wavelength retrieved in step 216, the sub-process 212 retrieves from the data repository 218 a list of potential interfering emission wavelengths associated with that emission wavelength and also stores the list of potential interfering emission wavelengths in the list 222.

The list of standard and interfering emission wavelengths 222 may be generated in any suitable structure or form, such as a look-up table containing a list of the top 10 emission wavelengths for each element in the list of elements and a set of potential interfering emission wavelengths for each emission wavelength.

The display information in FIG. 2c shows information available from the data repository 218. For example, for an element in the list of elements, such as manganese (Mn), the second order emission wavelength 224 of 259.372nm may encounter spectral interference from adjacent emission wavelengths of zirconium (Zr)226 at 259.371nm, molybdenum (Mo)228 at 259.371nm, iron (Fe)230 at 259.373nm, niobium (Nb)232 at 259.376nm, and so on.

The element search process 300 of the method 106 will now be described with reference to FIG. 3. The process 300 uses the data from the list 222 of standard emission wavelengths and interfering emission wavelengths loaded in the sub-process 212.

At query step 302, process 300 checks whether correlation process steps 400 through 500 for identifying spectral peaks and interference have been applied to each emission wavelength in list 222. If so, the process 300 proceeds to step 308. If not, process 300 proceeds to sub-process 400 for identifying spectral peaks in the sample spectral data. Query step 302 traverses each of the emission wavelengths in list 222 until all of the emission wavelengths in list 222 have been processed. The sub-process 400 will be explained in more detail below with reference to fig. 4, 5a and 5 d.

At query step 304, if a corresponding spectral peak in the sample spectral data has been identified with at least some confidence level, the process 300 proceeds to query step 306. If a spectral peak has been found, the corresponding emission wavelength is referred to herein as the analyte wavelength. A list of analyte wavelengths is compiled in the analyte wavelength list 310. If not, the process 300 returns to query 302 to locate the next emission wavelength in the list 222 for analysis.

At query step 306, process 300 determines whether the corresponding analyte wavelength is associated with a set of potentially interfering emission wavelengths based on data from list 222. If so, the process 300 proceeds to a sub-process 500 to identify spectral peaks of interfering emission wavelengths in the sample spectral data. If not, the process returns to query 302 to locate the next emission wavelength in list 222 for analysis. The sub-process 500 will be explained in more detail below with reference to fig. 6, 7a and 7 d.

At step 308, those analyte wavelengths identified as being associated with spectral peaks that may be affected by spectral interference are removed from list 310 as a result of process steps 400 through 500. Thus, list 310 is modified to ignore analyte wavelengths corresponding to spectral peaks that are determined to be likely to be affected by spectral interference (referred to herein as revised analyte wavelength list 312).

At subprocess 600, a set of rules is applied to the revised analyte wavelength list 312 to determine, with a certain level of confidence, the presence of one or more elements in the sample. The sub-process 600 will be explained in more detail below with reference to fig. 8. The revision list 312 may include a plurality of sub-lists. Each sublist includes a list of analyte wavelengths associated with a particular element.

A sub-process 400 for identifying spectral peaks for each analyte wavelength to compile the list 310 will now be described with reference to fig. 4.

At step 402, the sample spectral data from the data repository 210 is used for analysis. The sub-process 400 determines a region of interest of the sample spectral data corresponding to the respective analyte wavelength and sets the region of interest as the region of interest for the respective analyte wavelength.

At step 404, the sub-process 400 performs a constrained search within the region of interest to locate the relevant peak portion of the corresponding analyte wavelength. The search is constrained to prevent incorrect peak portions associated with adjacent spectral peaks from being incorrectly located.

At query step 406, the sub-process 400 determines whether the saturation intensity result is within the region of interest by determining whether a substantially flat portion is present at the upper end of the spectral peak. The saturation results include intensity measurements that exceed the measurable range of the instrument 102. If a substantially flat portion is located, the sub-process 400 proceeds to step 408, otherwise, the sub-process 400 proceeds to step 412.

At step 408, the sub-process 400 evaluates whether the located peak with a substantially flat portion is similar to a flat-top peak for the corresponding analyte wavelength (see, e.g., FIG. 5B).

At query step 410, if a flat-top peak is located in step 408, the sub-process 400 proceeds to step 416. Otherwise, the sub-process 400 for the respective analyte wavelength ends and an output indicating that no spectral peak corresponding to the analyte wavelength is located is provided as an input for step 304 of the process 300.

At step 416, the sub-process 400 determines that a spectral peak associated with the analyte wavelength has been found. However, due to the difficulty of accurately determining flat-top spectral peaks and saturation intensity results, a confidence rating indicating the level of confidence that a spectral peak corresponding to an analyte wavelength has been identified is given a low score. An output indicating that the spectral peak corresponding to the analyte wavelength has been located is provided as an input for step 304 of process 300.

At step 412, the sub-process 400 evaluates a general spectral peak (e.g., not a saturated result or a flat-top peak) of the region of interest in the sample spectral data. There may be more than one spectral peak in the region of interest.

At step 414, the sub-process 400 determines the appropriate local background location based on the peak portion located in step 404 to calculate the background standard deviation in step 420.

At query step 418, if step 414 determines that a spectral peak is present in the correlation peak portion, then the sub-process 400 proceeds to step 420. If not, the sub-process 400 ends. An output indicating that no spectral peak corresponding to the corresponding analyte wavelength is located is provided as an input for step 304 of process 300.

At step 420, a constrained search for local background measurements is performed near the spectral peak vertex position. Interpolation of background points allows determination of the net peak intensity at the peak location of the spectral peak. The Standard Deviation (SD) of the local background measurements of the spectral peaks was calculated. The local background with the largest SD is selected.

At step 422, a confidence factor based on the signal-to-noise ratio is assigned to the spectral peaks using the effective background SD and calculated as follows:

where the BG threshold is a scalar typically in the range of 1 to 10.

If the confidence factor passes the threshold test, then the spectral peak corresponding to the wavelength of the respective analyte is deemed detected, for example:

confidence factor > C threshold

Where the C threshold is a scalar typically in the range of 1 to 10.

If there is no valid background SD, the interpolated background value at the peak-to-peak position is replaced and the signal-to-background ratio is calculated instead of the signal-to-noise ratio.

Although the sub-process 400 above is described with reference to one example method of spectral peak detection, it should be understood that any suitable peak detection algorithm may be implemented. Some suitable example peak detection algorithms may include, but are not limited to, window thresholding methods, derivative analysis, and wavelet transforms.

Upon completion of step 422, an output indicating that the spectral peak corresponding to the respective analyte wavelength has been located is provided as an input for step 304 of process 300.

The displayed information in fig. 5a and 5b provide examples of flat top spectral peaks associated with the analyte wavelength of lithium (Li). As shown in FIG. 5a, the intensity of elemental Li in sample 1 was 2.4E +0mg/L (424). Fig. 5b shows that the analyte wavelength 670.783nm of Li was given a 1 star rating (reflecting a low confidence factor) due to the identification of a flat-top spectral peak 426 at 670.774nm around the analyte wavelength of 670.783 nm.

The displayed information in fig. 5c and 5d provide examples of conventional spectral peaks associated with analyte wavelengths of lithium (Li). As shown in fig. 5c, intensity 428 of elemental Li in sample 7 is 6.78mg/L fig. 5d shows that the analyte wavelength 670.783nm of Li is given a 5 star rating (reflecting a high confidence factor) due to the recognition of a conventional spectral peak 430 at 670.774nm around the analyte wavelength of 670.783 nm.

A sub-process 500 for identifying a spectral peak for each potential interfering emission wavelength identified in the query step 306 of the element search process 300 will now be described with reference to fig. 6.

At query step 502, a determination is made as to whether the subprocess 500 has evaluated all of the potentially interfering emission wavelengths identified in query step 306. If so, the sub-process 500 is complete. If not, the sub-process 500 obtains the next potentially interfering emission wavelength for evaluation and proceeds to step 504.

Based on the list of standard emission wavelengths and interfering emission wavelengths 222, the sub-process 500 selects a clean interfering emission wavelength to determine the presence of the relevant interfering element corresponding to the respective interfering emission wavelength at step 504. Typically, the sub-process 500 attempts to locate the most appropriate interfering emission wavelength from the list of emission wavelengths associated with the corresponding interfering element based on the list 222, which may produce a result with an acceptable confidence. For example, the sub-process 500 may select a interfering emission wavelength on the spectrum that is sufficiently separated from nearby emission wavelengths that may cause spectral interference with the interfering emission wavelength. Furthermore, the selected clean interfering emission wavelengths are not correlated with the saturation intensity results in the sample spectral data. Thus, the sub-process 500 attempts to locate a clean interfering emission wavelength that is least likely to be spectrally interfered with by itself, and may be capable of producing acceptable results.

At sub-process step 400, it is determined whether the selected clean interfering emission wavelength corresponds to a spectral peak in the sample spectral data using the same spectral peak identification method previously described with reference to FIG. 4.

If, at query step 506, the sub-process 400 determines that no spectral peak corresponds to a clean interfering emission wavelength, then the sub-process 500 determines that no interfering elements corresponding to the clean interfering emission wavelength are present in the sample and returns to query step 502 and obtains the next interfering emission wavelength for analysis. If the sub-process 400 determines that the spectral peak corresponds to a clean interfering emission wavelength, the sub-process 500 determines that an interfering element corresponding to the clean interfering emission wavelength is present in the sample, and the sub-process 500 proceeds to query step 508.

At query step 508, the sub-process 500 determines whether the detected interfering element is significant based on the confidence factor calculated in the sub-process 400. Typically, if the identified interfering element is associated with a confidence factor greater than a predetermined threshold in the range from 1 to 50, the sub-process 500 determines that the interfering element is significant and proceeds to step 510. If not, the sub-process 500 determines that the interfering element is not significant and returns to query 502 to retrieve the next available interfering emission wavelength.

At step 510, the distance between the analyte wavelength and the spectral peak of its associated interfering emission wavelength is determined. The measured and relative intensities of the detected interfering elements and their associated analyte elements are also determined. The measured intensity is used to calculate the ratio of interference to analyte(IAR or

) And the relative intensities are used to calculate the ratio of the relative intensity interference to the analyte (RIR or

)。

At query step 512, the sub-process 500 determines whether a potentially interfering emission wavelength should be determined as a likely interfering wavelength and added to the list of likely interfering emission wavelengths 516 based on the following three threshold tests:

peak separation < S threshold (1)

Wherein

hh is the maximum separation (typically in the range of 1.0 to 20.0) between the peaks of the spectral peaks corresponding to the analyte wavelength (analyte peak) and the associated interfering emission wavelength (interference peak), respectively,

hh is the minimum value (typically in the range of 0.1 to 10.0) of the ratio of the measured interference peak signal to the measured analyte peak signal,

hh is the minimum value (typically in the range of 1.0 to 20.0) of the ratio of the relative intensity of interference and the relative intensity of analyte.

The threshold test (1) determines whether the distance between the interference peak and the analyte peak is less than a threshold value in the range of 1 to 20. The threshold test (2) determines whether the IAR is above a threshold value in the range of 0.1 to 10.0. The threshold test (3) determines whether the RIR is above a threshold value in the range of 1.0 to 20.0.

If any of the above threshold tests are true, the sub-process proceeds to step 514 and the potentially interfering emission wavelengths are determined to be possible interfering wavelengths and added to the list 516. If not, the sub-process 512 returns to query 502 to retrieve the next available interfering emission wavelength.

For example, fig. 7a and 7b show that after analyzing sample 5(518, fig. 7a) using computer-implemented method 106, it has been identified that the confidence of detecting the analyte element bismuth (Bi) at an emission wavelength of 222.821nm is very low, because the confidence of detecting the interfering element chromium (Cr) at a nearby emission wavelength of 222.823nm is very high (520, fig. 7 b). In the spectrum shown in FIG. 7B, the presence of at least two adjacent spectral peaks near wavelength 222.821nm is also shown at region 522.

Similarly, fig. 7c and 7d show that after analyzing the sample 10 using the computer-implemented method 106 (524, 7c), it has been identified that the confidence of detecting the analyte element phosphorus (P) at the emission wavelength 213.618nm is very low, because the confidence of detecting the interfering element copper (Cu) at the nearby emission wavelength 213.598nm is very high (526, 7 d). In the spectrum shown in fig. 7d, the saturation result near the wavelength 223.619nm is also shown at region 528.

A sub-process 600 for determining, with a certain confidence level, the identification of one or more analyte elements in a sample will now be described with reference to fig. 8. The sub-process 600 is iteratively performed for each element in the list of elements to determine whether the element can be considered to be found in the sample based on the analyte wavelengths identified in the revised list 312 based on a set of predetermined criteria.

At query step 602, the sub-process 600 determines whether a minimum number of analyte wavelengths of the top 10 primary emission wavelengths are identified in the revised analyte wavelength list 312 for each element in the list of elements. In one embodiment, the sub-process 600 determines whether there is a minimum number of 2 analyte wavelengths for elements having more than two emission wavelengths and a minimum number of 1 analyte wavelength for elements having less than two emission wavelengths. Typically, the analyte wavelengths are ordered according to their associated confidence factors. If so, the sub-process 600 proceeds to query step 604. If not, the sub-process 600 proceeds to query step 622.

At query step 604, the sub-process 600 determines whether a minimum number of analyte wavelengths of the top 3 primary emission wavelengths are identified in the revised analyte wavelength list 312 for each element in the list of elements. In one embodiment, the sub-process 600 determines whether there is a minimum number of 2 analyte wavelengths for elements having more than two emission wavelengths and a minimum number of 1 analyte wavelength for elements having less than two emission wavelengths. Typically, the analyte wavelengths are ordered according to their associated confidence factors. If so, the sub-process 600 proceeds to query 606. If not, the sub-process 600 proceeds to query step 612.

At query step 606, if the minimum number of analyte wavelengths found is unlikely to be associated with any spectral interference, the sub-process 600 proceeds to step 608. Otherwise, for the currently evaluated analyte element, the sub-process 600 terminates, and the sub-process 600 is iteratively performed for the next element in the list of elements.

At step 608, the currently evaluated analyte element is considered present. The identified elements and their associated analyte wavelengths are added to the identified elements and analyte wavelengths list 110.

At query step 612, the sub-process 600 determines whether at least one analyte wavelength of the current analyte element is a strong dominant wavelength not affected by spectral interference (e.g., a confidence factor greater than 10) and whether at least one analyte wavelength of the current analyte element is a lower order analyte wavelength not affected by spectral interference from the first 10 dominant wavelengths (e.g., a confidence factor between 1 and 3). Essentially, at query step 612, the sub-process 600 determines whether there is at least one strong primary analyte wavelength and a weaker analyte wavelength supported for each element. If so, the sub-process 600 proceeds to step 608 and the current analyte element is considered to be found and added to the list 110 along with the associated analyte wavelength. If not, the sub-process 600 proceeds to query 614.

At step 614, the sub-process 600 checks whether lower order analyte wavelengths from the top 10 primary wavelengths are found for the evaluation element in the revised list 312. If so, the sub-process 600 proceeds to query 616. If not, the sub-process 600 terminates for the currently evaluated analyte element and the sub-process 600 is iteratively performed for the next element in the list of elements.

At query step 616, if all low-order analyte wavelengths found are spectrally disturbed, the sub-process 600 proceeds to step 620. If not, the sub-process 600 proceeds to step 618.

At step 618, the sub-process 600 checks whether some low order analyte wavelengths are found to have no spectral interference or relatively weak spectral interference. If so, the sub-process 600 proceeds to step 608 and the corresponding analyte element is deemed to be found at a lower order analyte wavelength. The element and associated analyte wavelength are then added to the list 110. If not, the sub-process 600 terminates for the currently evaluated analyte element and the sub-process 600 is iteratively performed for the next element in the list of elements.

At step 620, the sub-process 600 determines whether each of the discovered primary analyte wavelengths has a strong signal. If so, the sub-process 600 proceeds to step 608 and the corresponding analyte element is deemed to be found at a lower order analyte wavelength. The element and associated analyte wavelength are then added to the list 110. If not, the sub-process 600 terminates for the currently evaluated analyte element and the sub-process 600 is iteratively performed for the next element in the list of elements.

At query step 622, the sub-process 600 determines whether any strong analyte wavelengths have been found from the top 10 dominant wavelengths of the current analyte element (e.g., a confidence factor greater than 10). If so, the sub-process 600 proceeds to query step 624. If not, the sub-process 600 terminates for the currently evaluated analyte element, and the sub-process 600 is iteratively performed for the next element in the list of elements.

At step 624, the sub-process 600 determines whether any weaker analyte wavelengths have been found from the first 10 dominant wavelengths of the current analyte element (e.g., with a confidence factor between 1 and 3). If so, the sub-process 600 proceeds to query step 626. If not, the sub-process 600 proceeds to step 608 and the element is deemed to be found on the strong dominant wavelength identified in the query step 622. The element and associated analyte wavelength are then added to the list 110.

At step 626, the weaker analyte wavelengths (and the stronger primary analyte wavelengths from step 622) are deemed found for the analyte elements. The element and associated analyte wavelength are then added to the list 110.

Process 700 for reevaluating and verifying whether any analyte wavelengths associated with each element in list 110 are still spectrally disturbed. Thus, process 700 is a fine tuning step to re-evaluate each analyte wavelength of the elements in list 110 to remove any elements from list 110 that may be spectrally disturbed.

At query step 702, for each element in the list 110, the process 700 determines through process steps 704 through 714 whether any further corresponding analyte wavelengths are available for re-evaluation. If so, process 700 proceeds to step 704. If not, process 700 proceeds to sub-process 600.

At query step 704, the process 700 traverses all potential interfering emission wavelengths corresponding to the current analyte wavelength and selects the next available potential interfering emission wavelength for consideration at query step 706. If no more interfering emission wavelengths remain available, the process 700 returns to query 702. If there are more interfering emission wavelengths remaining available, the process 700 proceeds to a query step 706 for the next available interfering emission wavelength.

At query step 706, process 700 determines whether the current interfering transmission wavelength has a corresponding element in list 110. If so, process 700 proceeds to 708. If not, the process 700 returns to query 704 to retrieve the next available interfering wavelength.

At query step 708, if the interfering emission wavelengths have been previously identified as being associated with corresponding analyte wavelengths, the process 700 returns to query step 702 to retrieve the next analyte wavelength associated with the current element in the list 110. If the interfering emission wavelengths have not been previously identified as being associated with corresponding analyte wavelengths, process 700 proceeds to step 710 to determine the effect of the interference.

At step 710, a neighborhood scalar is computed to determine the importance of the disturbance. In general, the following calculation can be used, 1.0 minus the wavelength difference (in nm) between the analyte and the interference wavelength. Other suitable calculations based on the distance between the analyte wavelength and the corresponding interfering emission wavelength may also be used. In addition, the relative intensity ratio of the analyte wavelength peak intensity and the highest confidence element interference wavelength peak intensity will be calculated.

At query step 712, process 700 determines whether the proximity scalar exceeds a threshold (typically in the range of 0.2 to 1.0) and whether the scaled intensity and relative intensity are above a given threshold (typically in the range of 0.05 to 0.9). If so, process 700 proceeds to step 714. If not, the process 700 returns to query 702 to retrieve the next available analyte wavelength.

At query step 714, the process 700 updates the list of possible interfering emission wavelengths 516 with the interfering wavelengths evaluated in step 712.

In the sub-process 600, for each element in the list 110, once all corresponding analyte wavelengths of the identified element have been processed through steps 704 through 714, the sub-process 600 is re-executed based on the updated list of possible interfering emission wavelengths 516.

At query step 718, re-execution of the sub-process 600 determines whether each element in the current list 110 is present in the sample with an acceptable level of confidence. If it is determined with an acceptable level of confidence that the current element is not present in the sample, the process 700 ends. Otherwise, process 700 proceeds to step 720. In general, the sub-process 600 determines whether the analyte wavelength associated with the element is present with an acceptable level of confidence. It may then be inferred whether the element of interest is present in the sample based on the confidence level determined for the analyte wavelength.

At step 720, the process 700 removes the current element and associated analyte wavelength from the list 110.

A process 800 for evaluating analyte wavelengths from the list 110, generating a list of acceptable analyte wavelengths (ultimately for display by the display device 104) will now be described with reference to fig. 10. Typically, the selection is based on predetermined criteria, for example, including any one or more of:

whether the analyte wavelength correlates with a saturation result

Whether the analyte wavelength is related to spectral interference

Maximum number of analyte wavelengths displayed for each corresponding element

Whether the analyte wavelength is associated with a user selection

At query step 802, process 800 determines whether all elements in list 110 have been evaluated based on steps 804 through 812. If so, process 800 terminates. If not, process 800 proceeds to query step 804.

At query step 804, process 800 determines whether, for each element in list 110, the element is associated with at least one saturation intensity measurement of the analyte wavelength without spectral interference. If so, process 800 proceeds to step 806. If not, process 800 proceeds to step 808.

At step 806, the process 800 optionally includes a maximum of two saturated analyte wavelength measurements with the highest confidence factor in the list of acceptable analyte wavelengths.

At step 808, the process 800 selectively includes the unsaturated analyte wavelengths associated with the elements in the list 110 into a list of acceptable analyte wavelengths.

At query step 810, process 800 determines whether any user-selected wavelengths are already included in the list of acceptable analyte wavelengths. If so, process 800 returns to query step 802 and retrieves the next element from list 110 for processing. If not, the process 800 selectively includes the user-selected analyte wavelength in a list of acceptable analyte wavelengths. Typically, in such a case, the user-selected analyte wavelength is not determined to be the relevant analyte element that was considered to be found in a previously performed procedure. However, the results associated with the user-selected analyte wavelength will still be displayed in the list of acceptable analyte wavelengths.

A process 900 for examining the analyte wavelength, a list of acceptable analyte wavelengths for any anomalous results, to identify any total outliers that are typically the result of unrecorded interference (i.e., emission wavelengths that are not in the list of standard emission wavelengths and interfering emission wavelengths 222) will now be described with reference to fig. 11.

At step 902, the analyte wavelengths in the list of acceptable analyte wavelengths are sorted based on the measured concentration of the analyte element of interest in the sample. The concentration of each analyte element is determined based on the intensity concentration curve.

At query step 904, if there are more than two analyte wavelength results per element, process 900 proceeds to step 906. If not, process 900 proceeds to step 910.

At step 906, a quartering distance calculation is applied to the analyte wavelength. In other examples, one or more different calculations may be applied, such as a Z-score, a modified Z-score, a lognormal distribution, and so forth.

At step 908, for each analyte wavelength corresponding to an outlier result, a confidence factor associated with the analyte wavelength is decreased. Analyte wavelengths are also considered outliers.

At step 910, all analyte wavelengths are selected to remain in the accepted list of analyte wavelengths.

A process 1000 for selecting and ordering optimal analyte wavelengths for display on the display device 104 will now be described with reference to fig. 12. In process 1000, a threshold test is applied to the accepted analyte wavelength results from the outlier inspection process 900. To satisfy the threshold test, the analyte wavelength results must not correspond to saturation results and must have an acceptable calibration curve.

Metrics used to determine an acceptable calibration curve may include, but are not limited to, least squares goodness of fit correlation coefficients and percent Relative Standard Error (RSE). The calibration curve metric will be tested against an appropriate predetermined threshold. If the analyte wavelength results of the test are not met, a more relaxed test is used to include the calibrated analyte wavelength results. If no analyte wavelength results are calibrated, then all accepted analyte wavelength results are used. Finally, those analyte wavelength results that pass a given threshold test are ranked based on appropriate weights.

Examples of weighting factor calculations may include, but are not limited to:

confidence factor times the square root of the relative intensity of the analyte wavelength

Confidence factor of the wavelength of the analyte

The square root of the main order of the confidence factor divided by the wavelength of the analyte

The analyte wavelength result with the highest weight is selected to report the semi-quantitative concentration of the element.

At step 1002, the threshold test discussed above is applied. In particular, the process 1000 determines whether the analyte wavelength result corresponds to a saturation result and whether the analyte wavelength has an acceptable calibration curve (e.g., a relative standard error of less than 30%).

At query step 1004, if the analyte wavelength satisfies the threshold test (e.g., is not associated with a saturated result and has an acceptable calibration curve), process 1000 proceeds to step 1006. If not, process 1000 proceeds to step 1008.

At step 1006, the analyte wavelengths are ranked based on the weighting factor calculations described above. The analyte wavelength with the highest weight is selected for display on the display device 104 along with the associated concentration results for the corresponding analyte element.

At step 1008, a calibrated analyte wavelength test (e.g., a lower threshold test than in step 1002) is applied. For example, a boolean check is performed on the calibration curve to check whether the minimum criterion number is met. The calibration curve may also include acceptable correlation coefficients.

At query step 1010, if there is a calibrated analyte result based on the test in step 1008, then process 1000 proceeds to step 1006. If not, process 1000 proceeds to step 1012.

At step 1012, all analyte wavelengths in the list of acceptable analyte wavelengths are selected for use in the weighting factor calculation in step 1006.

Fig. 13 shows a graphical representation of results 1014 from a list of acceptable analyte wavelengths for display by display device 104. In particular, fig. 13 shows sample spectral data 1016 for sample 4. It is also shown that system 100 can select a portion 1018 of the spectrum for a detailed view in spectrum 1020, which provides an indicia of one or more analyte wavelengths and related analyte elements from a list of acceptable analyte wavelengths at corresponding locations on spectrum 1012.

In practice, ICP-OES techniques can be used to quantify up to 70 different elements in any given sample solution, with different samples likely containing different combinations and concentrations of these elements. The automatic element recognition functionality in embodiments of the present invention allows a user who is unaware of the contents of a solution to quickly identify the essential components present in the solution. This allows identification of unusual or unexpected sample components that may be ignored if a method of artificially identifying elements is employed.

Furthermore, the ICP-OES emission line of one element is often subject to spectral interference, which occurs when the analytical wavelength is partially or completely overlapped by the emission of another element or molecule, or is affected by unstructured background radiation. The presence and magnitude of spectral interference is highly dependent on the sample, even in the same method, and the appearance of analyte wavelengths affected by spectral interference may differ only slightly from analyte wavelengths that are not interfering; when the emission from the interferent is only slightly separated from the emission of the analyte wavelength, the presence of the interference may not be visually identified.

The interference avoidance function of embodiments of the present invention cross-references the multiple components of the spectral data for each measured solution to the known wavelength positions of all elements that can be quantified by ICP-OES. This allows for rapid and automatic identification of situations where there is interference at a given analyte wavelength, even in the case of complete spectral overlap between these interferents and the analyte wavelength. This allows the operator to identify the interference without knowing the potential spectral overlap or the contents of their solution.

An exemplary application of the system and method according to an exemplary embodiment of the present invention will now be described below.

Example 1: rapid sample evaluation and assisted method development for ICP-OES

In ICP-OES analysis, spectral interference and incorrect sample preparation are the two most common causes of erroneous results. Spectral interference varies widely between samples, especially in samples containing high concentrations of spectrally rich elements such as iron (Fe) or titanium (Ti). These disturbances may be overlooked, especially for inexperienced operators, and often appear in the reported results as abnormally high concentrations of the elements affected by the disturbance. Sample preparation errors are also difficult to detect and can have an effect on the results, depending on the particular error and preparation method used.

The system and method for ICP-OES according to embodiments of the present invention collects and interprets full spectrum data for each sample, adding only a few seconds to each analysis. The algorithm behind the interpretation will automatically identify the elemental composition of each sample, and the presence of spectral interference at common analyte wavelengths, without any input from the user. The following experimental results demonstrate the utility of the method and system in identifying significant spectral interference (see fig. 14) affecting several measurements in solid waste samples prepared according to standard method HJ781-2016, including lanthanum (La) on arsenic (As), iron (Fe) on manganese (Mn), and titanium (Ti) on vanadium (V). In each case, the interference and the suspected cause are clearly automatically flagged in the software user interface and an explicit rating system is used to represent the quality of the analyte peak at each wavelength of the analyte element. These experiments, which select challenging sample matrices, successfully demonstrated the robustness of spectral interference identification techniques, even in the most noisy spectra.

The interference information provided by embodiments of the present invention may be quickly and easily obtained without any method development or element selection on the part of the user. In some embodiments, the method 106 will automatically report the results for each element it detects in each sample on a per sample basis. Not only does a method need to be developed to obtain the information, but the information can remind a user of possible interference in the sample and give explicit suggestions to other wavelength qualities of each detection element, thereby providing a valuable first step for subsequent method development. The interference information may even be used to help select interference correction techniques, help ensure that these corrections are applied correctly, and compensate for the interference that is actually present in the sample being measured.

In some embodiments, identification of common sample preparation errors (insufficient hydrochloric acid added to the acid digest) has proven possible, as an additional function to the core function of high throughput screening applications. Semi-quantitative and convenient, real-time conditional formatting and filtering tools of chlorine (Cl) in samples can immediately identify samples with abnormally low levels of hydrochloric acid (see fig. 15). Advantageously, the user interface may ensure that even an inexperienced instrument operator can quickly and easily obtain such sample observations.

The user interface may provide a convenient set of graphical tools to display the contents of each measurement solution. In some embodiments, the user interface allows the user to obtain an immediate assessment of their sample content using a built-in color-coded periodic table thermogram graphic (see fig. 16a and 16 b). In general, the color coding of each element is customizable and correlated to the concentration of that element detected in each sample, allowing the user to make simple qualitative comparisons of the elemental content between samples (as also shown in fig. 15). In some embodiments, the visualization may be derived or included in the sample report after analysis.

Advantageously, the computer-implemented method 106 may be modular and compatible with other software modules to interface with the instrument 102. Embodiments of the present invention provide users with a high level of insight into their sample content while not requiring spectroscopic knowledge and requiring only minimal setup. Indeed, the elemental composition of the sample and all the information described in the preceding paragraph (including the appropriate sample absorption and rise delay) can be obtained within 15 seconds; screening of an entire 60 samples can be completed in only 15 minutes.

Example 2: simplified method development for extracting soil sample by DTPA

Soil samples were prepared for analysis according to the method of china HJ-804. Eight bioavailable elements in soil samples extracted by DTPA were determined using an Agilent 5800VDV ICP-OES equipped with an AVS 6 valve system and an SPS4 autosampler.

Sample screening is performed according to embodiments of the present invention and is used to aid in method development, resulting in high quality results and no re-measurement of samples. The reporting tool provided by embodiments of the present invention generates a quantitative worksheet to facilitate semi-quantitative analysis and provides sample insights to supplement quantitative data.

Method development can be tedious and time consuming. Imperfect methods can result in inaccurate data being reported and expensive re-measurements. Method development according to an example embodiment of the invention may include the following three steps.

Step 1: running sample

Sample screening according to method 106 is fast and easy to set up. It is not necessary to select any element or wavelength. The screening captures data for the entire wavelength range in approximately 15 seconds, and an automated element discovery algorithm selects elements and wavelengths for the operator.

Step 2: adding a recommended wavelength to a quantitative method

Screening presents a list of recommended wavelengths for each element detected in each sample.

For this application, all wavelengths selected by the screening process are also set forth in the HJ-804 specification method, indicating the reliability of the algorithm of method 106.

In the example of Mn (see fig. 17), the screening according to method 106 identified multiple wavelengths with a confidence rating of five stars, indicating that these wavelengths may be suitable for use in a quantitative method.

The output shown in fig. 17 suggests Mn 257.610 as the highest nominal analyte wavelength, based on the mass of the analyte peak and its immunity. The HJ 804 method recommends Mn 257.610 and Mn 293.305, corresponding to output analyte wavelengths associated with high confidence ratings.

The question mark next to the low star rated wavelength indicates that there are problems with the two main Mn lines. The pop-up box of the Mn 259.372 spectral line indicates very low confidence in the results due to strong Fe interference. The Mn 294.921 spectrum is also affected by Fe interference, as shown by method 106. Based on the knowledge of these samples, both wavelengths were excluded from the quantification method.

And step 3: operation quantitative method

Quantitative analysis was performed using the wavelengths recommended for the screening described above and semi-quantitative data was collected. This method allows you to run the prescribed method while also collecting semi-quantitative data for up to 70 elements in the sample, as shown in figure 18.

The same automated element discovery algorithm used for screening evaluated semi-quantitative data collected for each sample. The software calculates the approximate concentrations of all other elements in the sample and automatically identifies the presence of spectral interference.

The output data from the method 106 may be used to verify the overall quantitative result for a higher confidence in the result. As shown in the table below, the semi-quantitative concentration was within 25% of the full quantitative results, indicating that the results produced according to embodiments of the invention have satisfactory confidence.

Explanation of the invention

This specification, including the claims, is intended to be construed as follows:

the embodiments or examples described in the specification are intended to illustrate the invention, not to limit its scope. As the present invention can be embodied with various modifications and additions as readily appreciated by those skilled in the art, it is to be understood that the scope of the present invention is not limited to the exact construction and operation described or illustrated, but only by the appended claims.

The mere disclosure of method steps or product elements in the specification should not be construed as essential to the invention claimed herein unless explicitly stated or otherwise clearly stated in the claims.

The terms in the claims have the broadest meaning as is indicated by those of ordinary skill in the art at the time of the pertinent date.

The terms "a", "an" and "the" mean "one or more", unless expressly specified otherwise.

Neither the title nor the abstract of this application should be construed as limiting the scope of the claimed invention in any way.

When the objects, benefits or possible uses of the claimed invention are recited in the preamble of the claims, it does not limit the claimed invention to have only such objects, benefits or possible uses.

In the specification (including the claims), the term "comprise" and variations of the term (such as "comprises" or "comprising") are used to indicate "including but not limited to" unless the context or usage clearly dictates otherwise.

The disclosure of any document cited herein is incorporated by reference into this patent application as part of the disclosure, but is used for written description and implementation purposes only, and in no way should it be used to limit, define, or otherwise interpret any term of the application that does not provide a determinable meaning unless incorporated by reference. Any document incorporated by reference, itself, does not constitute an admission or approval of any statement, opinion or argument contained in any incorporated document.

The reference to any background or prior art in this specification is not an admission that such background or prior art forms part of the common general knowledge in the relevant art, or is acceptable prior art with respect to the validity of the claims.

Claims

1. A computer-implemented method of automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy, the method comprising the steps of:

obtaining sample spectral data from the sample,

obtaining, for each element of the periodic table of elements that is quantifiable by optical emission spectroscopy, a list of one or more predetermined emission wavelengths, each predetermined emission wavelength associated with a list of one or more potentially interfering emission wavelengths,

determining a list of one or more analyte wavelengths corresponding to spectral peaks in the sample spectral data based on the list of emission wavelengths,

determining, for each analyte wavelength, whether the corresponding spectral peak has a likelihood of being affected by an interfering emission wavelength that causes spectral interference based on the list of one or more potential interfering emission wavelengths corresponding to the analyte wavelength,

determining a revised list of one or more analyte wavelengths by removing from the list of analyte wavelengths that correspond to spectral peaks having a likelihood of being affected by interfering emission wavelengths, an

2. The computer-implemented method of claim 1, wherein

The sample spectral data includes data representing emission intensities corresponding to wavelengths within a sample spectral range, and

wherein the step of determining the list of analyte wavelengths comprises

Analyzing a region of interest of the spectral range of the sample corresponding to each predetermined emission wavelength of each element,

determining whether the saturation result is located within the region of interest,

determining whether a peak in the emission intensity is located within the region of interest upon determining that a saturation result is not located within the region of interest.

3. The computer-implemented method of claim 2, wherein the step of determining the list of analyte wavelengths further comprises determining whether the saturation result represents a peak in emission intensity with a flat top.

4. The computer-implemented method of claim 2 or 3, wherein the step of determining the list of analyte wavelengths further comprises

Determining a confidence level that a peak in the emission intensity has been identified in the region of interest based on a threshold test.

5. The computer-implemented method of claim 4, wherein determining a confidence level that a peak in the emission intensities has been identified in the region of interest comprises calculating a standard deviation of emission intensities near the peak to determine a confidence factor.

6. The computer-implemented method of claim 5, wherein an element associated with the peak is considered identified if the confidence factor is greater than a predetermined threshold.

7. The computer-implemented method of any of the preceding claims, wherein the step of determining whether the corresponding spectral peak for each analyte wavelength has a likelihood of being affected by an interfering emission wavelength comprises

8. The computer-implemented method of claim 7, wherein the step of determining a clean interfering emission wavelength comprises determining an interfering emission wavelength that is least likely to be affected by spectral interference.

9. The computer-implemented method of claim 7 or 8, further comprising:

10. The computer-implemented method of any of the preceding claims, wherein the set of criteria for determining a confidence level that one or more elements are present in the sample comprises any one or more of:

11. The computer-implemented method of claim 10, wherein

The first threshold is two for elements having at least three primary analyte wavelengths, and the first threshold is one for elements having two or less primary analyte wavelengths, and

the second threshold is at least one major analyte wavelength and one minor analyte wavelength.

12. The computer-implemented method of any of the preceding claims, further comprising

Based on the determined confidence level, one or more elements are added to the list of identified elements.

13. The computer-implemented method of claim 12, further comprising

14. The computer-implemented method of claim 12 or 13, further comprising

Selectively displaying analyte wavelengths corresponding to each element in the list of identified elements based on selection criteria, wherein the selection criteria include any one or more of:

whether the analyte wavelength is associated with a saturation result,

whether the analyte wavelength is associated with a user selection.

15. The computer-implemented method of any of claims 12 to 14, further comprising

Calculating the concentration of each element in the list of identified elements,

wherein the step of calculating the concentration of each element comprises measuring the emission intensity of a spectral peak associated with the corresponding element and correcting for background emission.

16. The computer-implemented method of any of the preceding claims, further comprising:

identifying abnormal analyte wavelengths, an

Reducing the confidence level that a corresponding element is present in the sample based on a measurement associated with the abnormal analyte wavelength.

17. A system for automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy, the system comprising:

a processor for performing the computer-implemented method of any of the preceding claims.

18. One or more tangible, non-transitory computer-readable media having computer-executable instructions for performing the computer-implemented method of any of the preceding claims.

19. A computer system for automatically identifying the presence of one or more elements in a sample via optical emission spectroscopy, the system comprising